Work

Comparative outcomes of endovascular versus synthetic bypass for isolated superficial femoral artery lesions: A 3-year single-center cohort study

Year: 2026

Type: article

Abstract: Peripheral artery disease (PAD) ranges from intermittent claudication to critical limb‑threatening ischemia (CLTI), with <3% of intermittent claudication cases progressing to CLTI. This study aimed to compare the outcomes of percutaneous transluminal angioplasty with stenting (PTAS) and synthetic graft femoropopliteal bypass (FPB) in a homogeneous PAD cohort. A subgroup analysis of TransAtlantic Inter-Society Consensus (TASC) B and C lesions, which lack standardized treatment algorithms, was also performed. We retrospectively analyzed 88 patients with no prior history of vascular interventions in the same limb who were treated for isolated superficial femoral artery lesions between 2016 and 2020 (29 underwent PTAS and 59 underwent FPB). Primary outcomes were reintervention, limb amputation, mortality, and vessel patency (primary, assisted primary, and secondary) over a 3‑year follow‑up. Subgroup analysis was performed to assess differences in treatment outcomes. No statistically significant differences were detected in reintervention, limb amputation, mortality, or patency between the two treatments. For loss of primary patency, the hazard ratio (HR) for FPB versus PTAS was 1.89 (95% confidence interval [CI], 0.76-4.69; log-rank p = 0.146) and 1.23 (95% CI, 0.45-3.38) after adjustment for lesion length and TASC grade. In the TASC B and C subgroup, no significant differences were observed in primary, assisted primary, or secondary patency (log-rank p = 0.809, 0.302, and 0.135, respectively). The study was underpowered to detect modest effects (post-hoc power for primary patency, 34.4%). In this small single-center cohort, no statistically significant difference in 3-year outcomes was detected between PTAS and FPB for isolated superficial femoral artery lesions; clinically meaningful differences cannot be excluded, and the findings should be regarded as hypothesis-generating.

Source: PLoS ONE

DOI: 10.1371/journal.pone.0361589